Lindahl Equilibrium as a Collective Choice Rule†

Faruk Gul and Wolfgang Pesendorfer

Princeton University

October 2021

Abstract

A collective choice problem is a finite set of social alternatives and a finite set of

economic agents with vNM utility functions. We associate a public goods economy with

each collective choice problem and establish the existence and efficiency of (equal income)

Lindahl equilibrium allocations. We interpret collective choice problems as cooperative

bargaining problems and define a set-valued solution concept, the equitable solution (ES).

We provide axioms that characterize ES and show that ES contains the Nash bargaining

solution. Our main result shows that the set of ES payoffs is the same a the set of Lindahl

equilibrium payoffs. We consider two applications: in the first, we show that in a large class

of matching problems without transfers the set of Lindahl equilibrium payoffs is the same as

the set of (equal income) Walrasian equilibrium payoffs. In our second application, we show

that in any discrete exchange economy without transfers every Walrasian equilibrium payoff

is a Lindahl equilibrium payoff of the corresponding collective choice market. Moreover,

for any cooperative bargaining problem, it is possible to define a set of commodities so that

the resulting economy’s utility possibility set is that bargaining problem and the resulting

economy’s set of Walrasian equilibrium payoffs is the same as the set of Lindahl equilibrium

payoffs of the corresponding collective choice market.

† This research was supported by grants from the National Science Foundation.

1. Introduction

Consider an organization that must decide among several alternatives that affect the

welfare of its members. Its goal is to find an equitable and efficient solution to the problem

when no transfers among members are possible. Examples include a community’s decision

on how to allocate infrastructure investments among neighborhoods, the allocation of office

space among groups within an organization, or the assignment of college roommates.

In this paper, we analyze the following mechanism for the allocation problems that

such an organization might confront: each group member is given an equal budget of fiat

money and confronts a price for each of the relevant alternatives under consideration. As

in standard consumer theory, members choose an alternative that maximizes their utility

subject to the budget constraint. The organization acts as an auctioneer and implements

an alternative that maximizes revenue. We allow choices to be stochastic; that is, agents

choose lotteries over social outcomes. We also allow the prices for the various social out-

comes to be individual-specific. Thus, our mechanism is analogous to a Lindahl mechanism

(Lindahl (1919)) in a public goods setting. It differs from the standard Lindahl equilibrium

in that the objects of choice are social outcomes (allocations) rather than individual con-

sumptions. We refer to this mechanism as a collective choice market and to its equilibria

as Lindahl equilibria.

As an example, suppose a town with 2n inhabitants considers implementing project A

or project B. The status quo yields zero utility for all inhabitants who are divided equally

between A and B-supporters. Each inhabitant receives utility 1 if her favored alternative

is implemented and zero otherwise. In the collective choice market, every agent is given

one unit of fiat money and prices for the two alternatives; she must choose an optimal

lottery over the two alternatives subject to her budget constraint. In the unique Lindahl

equilibrium, the n agents who prefer A must pay 2 for alternative A and zero for B; B-

supporters must pay 2 for B and zero for A. At these prices, it is optimal for all agents to

choose the lottery that yields A and B with equal probabilities. This lottery also maximizes

auctioneer revenue and is thus a Lindahl equilibrium.

The town’s decision problem can also be described as an n-person bargaining problem

in which the attainable utility profiles correspond to the inhabitants’ utilities for lotteries

1

over A and B. More generally, every collective choice market can be mapped to an n-

person bargaining problem. We introduce a new solution concept for n-person bargaining

problems - the equitable solution. Unlike standard solution concepts such as the Nash

bargaining solution or the Kalai-Smorodinsky solution, the equitable solution is set valued.

That is, it associates with every bargaining problem a set of equitable solutions. Our main

result (Theorem 2) relates Lindahl equilibrium outcomes and equitable outcomes of the

associated bargaining game.

To define the equitable solution, we first identify a subclass of bargaining problems

that have a self-evidently fair outcome. In those bargaining problems agents have linear

utility over a fixed surplus to be divided among them and all standard bargaining solutions

agree that equal division is the only plausible equitable outcome.1 For a general bargaining

problem; that is, when a fair solution is not self-evident, the equitable solution picks

all those outcomes which are the obvious fair outcome of some dominating bargaining

problem.2 In Theorem 1 we provide an axiomatic foundation for the equitable solution.

The equitable solution corresponds to a broad notion of equity. As the following

example illustrates, it is often the case that several solutions to a bargaining game may

qualify as equitable. Suppose Ann and Bob must divide two cakes, a peanut butter cake

and a chocolate cake. Utilities are linear but Bob is allergic to peanuts while Ann likes

the peanut butter cake just as much as the chocolate cake. One equitable division might

be to give Ann the peanut butter cake and divide the chocolate cake equally between

Ann and Bob. To justify this outcome, Ann could argue as follows: Since Bob has no

use for the peanut butter cake, from his perspective the situation is as if we only had the

chocolate cake and in that situation it’s obviously equitable to divide the chocolate cake

equally. Perles and Maschler (1981) provide an axiomatic foundation for Ann’s argument.

The Perles-Maschler solution to the bargaining game would give Ann the peanut butter

cake and divide the chocolate cake equally between the two players. On the other hand,

Bob could make the following argument: If I were not allergic to peanuts we would each get

1 See Thomson (1994) for a survey of bargaining solutions. For our setting, the relevant bargainingsolutions are ordinal, that is, solutions that are invariant to affine transformation of utilities. The Nashbargaining solution (Nash (1950)), the Kalai-Smorodinsky solution (Kalai and Smorodinsky (1975)) andthe Perles-Maschler solution (Perles and Maschler (1981)) are examples of ordinal solutions.

2 Bargaining problem A dominates problem B if for every utility profile in B there is a better utilityprofile in A and for every utility profile in A there is a worse utility profile in B.

2

one cake. Since Ann is indifferent between the chocolate cake and the peanut butter cake it

makes sense that she gets the peanut butter cake and I get the chocolate cake. Nash (1951)

provides an axiomatic foundation for Bob’s argument. In the Nash bargaining solution of

this game, Ann gets the peanut butter cake and Bob gets the chocolate cake. Of course,

in-between solutions can also be justified. For example, the Kalai-Smorodinsky solution

suggests giving 2/3 of the chocolate cake to Bob and 1/3 to Ann while Ann gets all of the

peanut butter cake.

The cake division problem can be mapped to a collective choice market if we interpret

shares of the cake as probabilities of getting the cake. An allocation specifies who receives

which cake. There are two Pareto efficient allocations for the utilities specified above:

allocation (a) gives both cakes to Ann while allocation (b) gives Ann the peanut butter

cake and Bob the chocolate cake. In a collective choice market, Ann and Bob each have a

budget of 1 unit of fiat money and confront prices for allocations (a) and (b). Allocation

(b) corresponds to the Nash bargaining solution and it can be supported as a Lindahl

equilibrium with the following prices: Bob must pay 1 for (b) and 0 for (a) while Ann

must pay 1 for (b) and 2 for (a). Choosing (b) with certainty maximizes the utilities of

both agents at these prices and, since both allocations yield the same revenue, (b) is an

optimal choice for the auctioneer as well. The Perles-Maschler solution corresponds to

the allocation that yields (a) and (b) with equal probabilities. It, too, can be supported

as an equilibrium of the collective choice market. To do so, we set Bob’s prices to 0 for

(a) and 2 for (b) and we set Ann’s prices to 2 for (a) and 0 for (b). At these prices, the

Perles-Maschler allocation maximizes each agents utility and the auctioneer’s revenue. This

example illustrates how the equilibria of a collective choice market replicate the outcomes

of the equitable solution. Our Theorem 2 shows that this is no accident.

Theorem 2 is our main result and it shows that the set of Lindahl equilibrium payoffs

coincides with the equitable outcomes of the corresponding bargaining game. One direc-

tion, every Lindahl equilibrium is equitable, shows that our axioms identify the normative

properties of collective choice markets. The second direction, every Lindahl equilibrium is

equitable, shows that collective choice markets are flexible enough to implement any equi-

table outcome. This direction is reminiscent of the second theorem of welfare economics.

3

We first apply our results to matching. A group of individuals must decide who

matches with whom as, for example, in a roommate assignment problem or in the clas-

sic two-sided matching problem (Gale and Shapley (1962)). We show that in matching

problems without transfers the Lindahl equilibrium outcomes of a collective choice market

coincide with Walrasian equilibrium outcomes of market for partners. In a collective choice

market, consumers express their demands for social outcomes (that is, allocations) while

in a Walrasian equilibrium, consumers express their demand for potential partners. In

both cases, utility is non-transferable and consumers have the same budget of fiat money.

Our result implies that the Walrasian equilibria of matching markets with non-transferable

utility and equal budgets are equitable.

In our second application, we consider the allocation of indivisible goods when con-

sumers have identical budgets of fiat money. Every Walrasian equilibrium (with combi-

natorial prices) of this exchange economy is a Lindahl equilibrium of the corresponding

collective choice market. However, the set of Walrasian equilibria is sensitive to the exact

specification of the traded goods (that is, how property rights are defined) while Lindahl

equilibria are not. More precisely, the set of Lindahl equilibrium payoffs for two collective

choice markets is the same whenever the two choice problems lead to the same bargaining

set (i.e., the same set of feasible utility vectors). In contrast, the set of Walrasian equi-

librium payoffs may be different for two different exchange economies that yield the same

bargaining set. We show that every possible bargaining set can be “commodified” as an

exchange economy; that is, it is possible to define an exchange economy that yields the

particular bargaining set as its set of feasible utility vectors. Moreover, this commodifica-

tion can be done in a manner that renders the set of Walrasian equilibrium payoffs equal

to the set of Lindahl equilibrium payoffs. Finally, we show that for 2-person economies,

a commodification that renders the set of Walrasian equilibrium payoffs identical to the

Lindahl equilibrium payoffs can be done with additive preferences and additive prices.

1.1 Related Literature

Our paper is related to the extensive literature on axiomatic bargaining theory (see

Thomson (1994) for a survey). Our axiomatic treatment is related to Nash (1950) and

4

we discuss this relationship in detail after the statement of Theorem 1 below. For 2-

person bargaining, our solution concept includes the Kalai-Smorodinsky solution (Kalai

and Smorodinsky (1975)) and the Perles-Maschler solution (Perles and Maschler (1981))

in the set of equitable outcomes.

Hylland and Zeckhauser (1979) were the first to propose Walrasian equilibria as so-

lutions to stochastic allocation problems in situations without transfers. Gul, Pesendorfer

and Zhang (2020) extend Hylland and Zeckhauser from unit demand preferences to gen-

eral gross-substitutes preferences. Collective choice markets allow for arbitrary preferences,

public goods and externalities and hence provide a further generalization of the environ-

ment considered in these papers.

Our Corollary 1 is related to research that analyzes the Nash product rule, that is,

the allocation rule that maximizes the product of utilities. Eisenberg and Gale (1959)

show that competitive equilibria of a Fisher market with equal budgets yield the Nash

product rule. Fain, Goel and Munagala (2016) consider an economy in which the planner

must allocate a fixed budget among a finite collection of public goods with linear costs and

utilities. They show that the Nash product rule yields a Lindahl equilibrium allocation of

an economy in which each agent controls an equal share of the budget. Our Corollary 1 also

establishes a connection between Lindahl equilibria and the Nash product. In contrast to

Fain, Goel and Munagala (2016), we do not assume linear costs3 and, as a result, typically

there are multiple Lindahl equilibria. Nonetheless, we show that each Lindahl equilibrium

implements the Nash product rule for appropriately modified utilities. Brandl, Brandt,

Peters, Stricker and Suksompong (2020) examine the incentive properties of the Nash

product rule and show that it satisfies a weak form of incentive compatibility.

Foley (1967), Schmeidler and Vind (1972) and Varian (1974) associate equity with

envy-freeness. Walrasian equilibria with equal budgets are envy free and, thus, these

authors establish a connection between competitive outcomes and equity. In a public

goods setting, two agents may contribute different amounts to the same public good and

hence it is not straightforward to adapt the notion of envy freeness to Lindahl equilibria.

Moreover, as we show in section 5.2, there are multiple ways to commodify each collective

3 Our model allows an arbitrary, but finite, set of feasible allocations.

5

choice market. The same utility profile may be envy free for one specification but fail envy-

freeness for another. Thus, collective choice markets do not lend themselves to a coherent

definition of envy-freeness. Instead, we provide a definition of equitable outcomes based

on the associated cooperative bargaining problem and show its equivalence to Lindahl

equilibria.4

2. Collective Choice Markets and Lindahl Equilibrium

Consider the following collective choice problem: n agents, i ∈ {1, . . . , n}, must decide

on one of k social outcomes, j ∈ K = {1, . . . , k}. A random outcome, q ∈ Q := {q ∈

IRk+ |∑j∈K q

j = 1}, is a probability distribution over the social outcomes. Agents are

expected utility maximizers; i’s utility if outcome j ∈ K occurs is uji ≥ 0 and ui =

(u1i , . . . , u

ki ) denotes i’s utility index. In addition to the k outcomes described above, there

is a disagreement outcome that yields zero utility to every agent. We dismiss all agents

who have no stake in the collective decision; that is, we assume that for every utility ui

there is some j such that uji > 0. The vector u = (u1, . . . , un) denotes a profile of utilities.

We will define a social choice rule by identifying the Lindahl equilibria of a corre-

sponding market economy, the collective choice market. This market has n + 1 agents,

the n described above, now called consumers, and one firm. Consumer i has one unit

of fiat money and can purchase quantity qj ≥ 0 of each “public good” j ∈ {1, . . . , k} at

price pji ≥ 0. Outcome 0 is identified with not purchasing any good (and therefore has

price 0). Let e = (1, . . . , 1) be the k−dimensional unit vector. Consumer i faces prices

pi = (p1i , . . . , p

ki ) and solves the following maximization problem:

Ui(p) = maxqui · q subject to pi · q ≤ 1, e · q ≤ 1 (1)

We say that q is an minimal-cost solution to the maximization problem above if q solves

that problem and p · q ≤ p · q for every other solution q. The price paid for outcome j

depends on the identity of the consumer. We let pi = (p1i , . . . , p

ki ) ∈ IRk+ be consumer

4 Sato (1987) adapts the notion of envy-freeness to Lindahl equilibria by assuming that agent i convertsj’s actual consumption of the public good into a virtual quantity based on j’s utility of that good. In effect,Sato identifies a commodity space and associated utility functions for which envy freeness of Lindahlequilibria with equal budgets is satisfied.

6

i’s prices and p = (p1, . . . , pn) ∈ IRkn+ be a price profile. The firm chooses q to maximize

profit, that is, to solve

R(p) = maxq

n∑i=1

pi · q subject to e · q = 1 (2)

Definition: The pair (p, q) is a Lindahl equilibrium (LE) for the collective choice market

if q is a minimal-cost solution to every consumer’s maximization problem at prices pi and

solves the firm’s maximization problem at prices p.

As is well known, a Lindahl equilibrium can be re-interpreted as a Walrasian equilib-

rium of a suitably modified alternative economy. To do so, we identify i’s consumption of

public good j with a distinct private good (i, j) and assume the firm can meet demand

(q1j . . . . , qnj) with supply q if max1≤i≤n qij ≤ qj for all j. This connection to Walrasian

equilibrium means that existence of a Lindahl equilibrium can be shown by adapting stan-

dard results. The proof of Lemma 1 and subsequent results are in the appendix.

Lemma 1: Every collective choice market has a Lindahl equilibrium; all Lindahl equi-

libria are Pareto efficient.

A broad range of applications including all discrete allocation and matching problems

can be modeled as collective choice markets. For example, if the aggregate endowment

consists of a collection of indivisible goods, then the set of outcomes is simply the set of

all allocations.

Similarly, to map a two-sided matching problem into a collective choice market, we

partition “consumers” into two groups, N1, N2 and let the outcomes be the set of all

matchings; that is, one-to-one functions j : N1 ∪N2 → N1 ∪N2 such that j(j(i)) = i and

[i ∈ Nl implies j(i) = i or j(i) /∈ Nl]. Hence, a member of group l is either unmatched

(j(i) = i) or is matched with someone from the other group (j(i) /∈ Nl). In section 4, we

analyze these applications and relate Lindahl equilibrium outcomes to standard Walrasian

outcomes of these economies.

Let L(u) ⊂ IRn+ be the set of utility profiles that can be supported as a LE of the corre-

sponding collective choice market u; that is, L(u) = {(u1 ·q, . . . , un ·q) | (p, q) is a LE of u}.

7

3. The Bargaining Problem

In this section, we define and characterize the equitable solution, a cooperative solution

concept. For any x, y ∈ IRn, we write x ≤ y to mean xi ≤ yi all i. For any bounded set

X, we let d(X) (the disagreement point of X) be the greatest lower bound of X and we

let b(X) (the bliss point X) be the least upper bound of X.

For any finite, bounded set X ⊂ IRn, let convX denote its convex hull and compX

denote its comprehensive hull of X; that is,

compX := {x ∈ IRn | d(X) ≤ x ≤ y for some y ∈ X}

Then, the set cocoX := comp convX is it convex comprehensive hull; that is, the smallest

(in terms of set inclusion) convex and comprehensive set that contains X.

For any a, x ∈ IRn, let a ⊗ x = (a1 · x1, . . . , an · xn), a ⊗ B = {a ⊗ x |x ∈ B} and

B + z = {x + z |x ∈ B}. Let ei denote unit vector with zeros in every coordinate except

i, o := (0, . . . , 0) ∈ IRn, e := (1, 1, . . . , 1) ∈ IRn and ∆ = conv {o, e1, . . . , en}.

We consider bargaining problems in which the set of attainable utility profiles is a

full dimensional, comprehensive polytope. Let B denote the set of all such polytopes and

Bo = {B ∈ B | o = d(B)} denote the set of all polytopes with disagreement point o;

that is, the set of normalized polytopes. A (set-valued) bargaining solution is a mapping

S : B → 2IRn\∅ such that S(B) ⊂ B. Hence, a bargaining solution chooses a nonempty set

of alternatives from every bargaining problem. Below, we define a set-valued bargaining

solution, which we call the equitable solution (ES) and provide axioms that characterize it.

Equal division is the natural fair outcome if the bargaining set is the unit simplex.

Since von Neumann-Morgenstern utilities are unique only up to positive affine transforma-

tions, any affine transformation of the unit simplex should yield the corresponding affine

transformation of the fair outcome. Let D be the bargaining problems that are positive

affine transformations of the unit simplex, that is, B ∈ D ⊂ B if there exists a, z ∈ IRn

such that ai > 0 for all i and B = a ⊗∆ + z. For B = a ⊗∆ + z ∈ D the fair outcome

is x = 1na ⊗ e + z. Hence, only bargaining problems that are equivalent, up to a positive

affine transformation of utilities, to ∆ have fair outcomes, and the fair outcome is the

corresponding affine transformation of to the unique symmetric and efficient outcome of

8

∆. Let F (B) be the set of fair outcomes of B ∈ B with the convention that F (B) = ∅ if

B 6∈ D.

Definition: Let A,B ∈ B. Then, A ≤ B if for every x ∈ A, y ∈ B, there exist x′ ∈ A,

y′ ∈ B such that x′ ≤ y and x ≤ y′.

Thus, A ≤ B if for every utility vector is in B there is a corresponding utility vector

in A that dominates it and, conversely, for every utility vector in A there is a correspond-

ing utility vector in B that is dominated by it. The equitable solution consists of those

outcomes of B that coincide with the fair outcome of a simplex A such that B ≤ A.

Definition: The equitable solution is E(B) := {x ∈ A ∩ F (B) |B ≤ A}.

With some abuse of terminology, we also refer to the mapping from bargaining prob-

lems to their equitable sets as ES. Below, we provide axioms on the bargaining solution

S that characterize the equitable solution. The first axiom, scale-invariance, is familiar

from other bargaining solutions including the Nash bargaining solution and the Kalai-

Smorodinsky solution. It asserts that positive affine transformations of utilities do not

change the set of chosen (physical) outcomes.

Scale Invariance: S(a⊗B + z) = a⊗ S(B) + z whenever ai > 0 for all i.

Our symmetry axiom applies only to the bargaining problem ∆. It ensures that the

unique symmetric and efficient outcome of ∆ is the only one chosen.

Symmetry: S(∆) = { 1n · e}.

The following axiom is similar to independence of irrelevant alternatives (IIA) of the

Nash bargaining solution.

Consistency: B ≤ A implies S(A) ∩B ⊂ S(B).

One important difference between consistency and IIA is that the former is applicable

even if d(A) 6= d(B). The latter replaces ≤ with ⊂ and requires d(A) = d(B). The

next axiom, justifiability, is our main assumption. We discuss its interpretation following

Theorem 1 below.

Justifiability: x ∈ S(B) implies B ≤ A and {x} = S(A) for some A.

9

Theorem 1 below reveals that the preceding four axioms characterize ES.

Theorem 1: The only bargaining solution that satisfies the four axioms above is ES.

It is easy to verify that E satisfies the axioms. For the converse, let S be a bargaining

solution that satisfies the axioms and note that scale-invariance and symmetry imply that

S(a⊗∆ + z) = 1n{a⊗ e + z} = F (a⊗∆ + z) for all a ∈ IRn++ and z. Then, consistency,

ensures E(B) ⊂ S(B) while justifiability implies x /∈ S(B) whenever x /∈ E(B) for all B.

To see how our axioms relate to the axioms for the Nash bargaining solution (Nash

1950), consider a single-valued solution. In that case, adding d(A) = d(B) to consistency

and justifiability implies that the Nash bargaining solution is characterized by either of

these two axioms together with scale invariance and symmetry. We can get this result

even if we replace ≤ with ⊂ while maintaining d(A) = d(B) in both consistency and

justifiability.

The equitable solution is a permissive solution concept that, in the case of two agents,

includes all the standard scale invariant bargaining solutions. In particular, the Nash bar-

gaining solution (Nash (1950)), the Kalai-Smorodinsky solution (Kalai and Smorodinsky

(1975)) and the Perles-Maschler solution (Perles and Maschler (1981)) are contained in the

equitable solution when n = 2.

To see the motivation for justifiability consider again the example in the introduction:

Ann and Bob must divide two cakes, a peanut butter cake and a chocolate cake. Utilities

are linear but Bob is allergic to peanuts while Ann likes the peanut butter cake just as

much as the chocolate cake. The corresponding bargaining set B is the convex hull of the

points {(0, 0), (1, 0), (1/2, 1), (0, 1)}. Ann could argue as follows: I should get the peanut

butter cake since you don’t have any use for it. We both like the chocolate cake and so

it makes sense to divide it equally. So, we should choose x = (3/4, 1/2) as our solution.

Ann’s argument leads to the Perles-Maschler solution of this bargaining game. On the

other hand, Bob could make the following argument: If I were not allergic to peanuts we

would each get one cake. Since you are indifferent between the chocolate and the peanut

butter cakes it makes sense that you get the peanut butter cake and I get the chocolate

cake. So we should choose y = (1/2, 1) as our solution. Bob’s argument leads to the

Nash-bargaining solution of this bargaining game.

10

Justifiability says that both Ann and Bob have valid arguments and, thus, permits

both solutions as possible outcomes. More precisely, Ann can justify her favored outcome

by noting that the bargaining game A = conv{(1/2, 0), (1, 0), (1/2, 1)} is a translation of

a simplex and its fair outcome is x. Since B ≤ A; that is, A dominates B, her proposed

alternative x is justified. Bob can justify his favored outcome by pointing to the bargaining

game A′ = conv{(0, 0), (1, 0), (0, 2)} which also dominates B and has y as its fair outcome.

The following lemma characterizes the equitable solution for the 2-player case and

yields as a straightforward corollary the fact that the three solutions above are contained

in the equitable solution.

Lemma 2: If n = 2, then E(B) consists of all Pareto efficient utility profiles x ∈ B such

that d(B) + 1/2(b(B)− d(B)) ≤ x.

Lemma 2 shows that in the case of two agents the equitable solution consists of all

Pareto efficient outcomes that satisfy midpoint domination. The midpoint corresponds to

the outcome that would be achieved if a coin is flipped to decide which agent is made a

dictator. In his survey of bargaining solutions, Thomson (1994, p. 1254) points out that

most bargaining solutions satisfy midpoint domination. In particular, the three solutions

mentioned above satisfy it.

As Lemma 2 shows, the two agent case allows a simple characterization of the set of

equitable outcomes. However, when there are more than 2 agents this is no longer the

case. In particular, while the equitable solution continues to satisfy midpoint domination,

the converse is no longer true. Not every Pareto efficient outcome that satisfies midpoint

domination can be equitable. For example consider the 3-person bargaining problem B =

coco{(0, 0, 0), (1, 1, 1/2), (1/3, 1/3, 1)}. Then, x = (1/3, 1/3, 1) is a Pareto efficient outcome

that satisfies midpoint domination (since each agent receives at least 1/3) but it is not an

equitable solution since there is no A ∈ D, B ≤ A that has x as its fair outcome. Our main

result, Theorem 2 below, characterizes the set of equitable outcomes as the set of Lindahl

equilibrium outcomes of the corresponding collective choice market.

Before doing so, we relate the equitable solution to the Nash bargaining solution. For

any B ∈ B, let

fB(x) :=∑i

log(xi − di(B))

11

The Nash bargaining solution of B, η(B), is the unique outcome that maximizes fB(·) in

B.

Definition: The outcome x ∈ B is Nash-sustainable if there exists A ∈ B, B ≤ A such

that x = η(A).

Note that the fair outcome coincides with the Nash bargaining solution if the bar-

gaining set is a positive affine transformation of a simplex. Therefore, any outcome in the

equitable solution must be Nash sustainable. Lemma 3, below, shows that the converse is

true as well. We let N(B) denote the set of Nash-sustainable outcomes of the bargaining

problem B.

Lemma 3: For all B ∈ B, E(B) = N(B).

Lemma 3 shows that Nash sustainability is an alternative characterization of the

outcome function that satisfies our axioms.

4. The Equitable Solution and Lindahl Equilibria

Any collective choice problem, u, corresponds to a unique bargaining problem

Bu := coco ({(uj1, . . . , ujn) | j ∈ K} ∪ {o})

Hence, Bu is the convex and comprehensive hull of the utilities in the collective choice

problem. Our main result, Theorem 2 below, shows that the equitable outcomes of Bu

coincide with the Lindahl equilibrium payoffs of the collective choice problem u.

One direction, every Lindahl equilibrium is equitable, shows that the equitable so-

lution captures the notion of fairness embodied in Lindahl equilibria of collective choice

markets. Just as the first welfare theorem clarifies the notion of efficiency embodied in

competitive models, Theorem 2 shows that Lindahl equilibria with equal budgets represent

outcomes that are justifiable in the sense of our final axiom above. The other direction,

every equitable solution is a Lindahl equilibrium, shows that all equitable solutions can be

implemented via a collective choice market. Thus, this direction of Theorem 2 is analogous

to the second welfare theorem albeit without the need to vary consumer’s wealth.

12

Theorem 2: The set of Lindahl equilibrium payoffs is the same as the equitable solution

of the corresponding bargaining problem: L(u) = E(Bu) for all u.

To see how every equitable outcome can be made into a Lindahl equilibrium payoff,

take any equitable a ∈ Bu. By definition, there exists some bargaining set A that (i)

is a positive affine transformation of a simplex; (ii) has a as its fair outcome; and (iii)

satisfies Bu ≤ A. Clearly, a must be on the Pareto-frontier of Bu. In general, d(A) need

not be the origin but in the special case where it is we have Bu ⊂ A and A = na ⊗ ∆.

Consider that special case. Since a is Pareto efficient, there is a distribution q over Pareto

efficient outcomes that delivers the utility vector a. The vector q is the Lindahl equilibrium

allocation. To find the Lindahl equilibrium prices, let yj = (uj1, . . . , ujn) be the utility vector

corresponding to outcome j. Since Bu ⊂ A, the vector y must be a convex combination

of the extreme points of A = na ⊗∆. Hence, there are zji ≥ 0 such that∑i zji = 1 and∑

i naizji · ei = yj for all j. Then, set pji = nzji to be the price agent i pays for outcome

j. It is then not difficult to verify that (p, q) is a Lindahl equilibrium. Our proof that

E(Bu) ⊂ L(u) for all u generalizes this argument to cover the case in which d(A) 6= o; that

is, when we have Bu ≤ A but not Bu ⊂ A.

For the converse; that is, to see that every Lindahl equilibrium payoff is equitable, let

(p, q) be a Lindahl equilibrium and recall that q solves the linear program below for every

consumer i:

maxq′

ui · q′ subject to pi · q′ ≤ 1, e · q′ ≤ 1 (1)

Let αi be the shadow price of the constraint pi ·q′ ≤ 1 and let ci be the shadow price of the

constraint e · q′ ≤ 1. Then, the constraint of the dual of the above linear program requires

that for all i, j,

αipji ≥ u

ji − ci (3)

Moreover, optimality of q implies that inequality (3) holds with equality if qj > 0. Suppose

αi, the shadow price of the budget constraint, is strictly positive for every consumer. Then,

pi · q = 1 and, since αipji = uji − ci if qj > 0, we have αi + ci = ui · q for all i; that is,

α+ c, where α = (α1, . . . , αn) and c = (c1, . . . , cn), is the payoff vector associated with the

equilibrium (p, q). Note that α+ c is the fair outcome of the bargaining game nα⊗∆ + c.

13

The key step of the proof establishes that Bu ≤ nα⊗∆ + c which then implies that α+ c

is an equitable outcome of Bu.

Theorem 2 shows that Lindahl equilibria are equitable, which, by Lemma 3, implies

that every Lindahl equilibrium is Nash sustainable. Thus, if Bu is the bargaining game

corresponding to the collective choice problem then each Lindahl equilibrium payoff is

the payoff in the Nash bargaining solution of a bargaining game A such that Bu ≤ A.

In Corollary 1, below, we rely on this connection to characterize all Lindahl equilibrium

allocations in terms of the Nash bargaining solution.

Call an allocation q a Nash allocation for u if it yields the Nash bargaining solution

payoffs for the associated bargaining problem Bu; that is, if u · q = η(Bu). Note that while

the Nash bargain solution, η(Bu), is a unique element of Bu, there may be multiple Nash

allocations. Each one of these allocations solves the following maximization problem:

N∗ = maxq∈Q

∑i

log(ui · q) (N)

To relate Lindahl equilibria to Nash allocations, we define a family of utilities derived

from the original utility u of the collective choice problem. Let Ci(ui) = [0,maxj uji ) and

C(u) =∏ni=1 Ci(ui). For ci ∈ Ci(ui), define

uji (ci) := max{uji − ci, 0}

Since c ∈ C(u), uji (ci) > 0 for some j and hence ui(ci) = (u1i , . . . , u

ki ) is a utility. Let

u(c) = (u1(c1), . . . , un(cn)) be the corresponding profile of utilities.

Corollary 1: If q is the Nash allocation for u(c), c ∈ C(u) and uji ≥ ci whenever qj > 0,

then (p, q) such that pi = ui(ci)/(ui(ci) · q) is a Lindahl equilibrium for u. Conversely, if

q is a Lindahl allocation for u, then there is c ∈ C(u) such that q is a Nash allocation for

u(c) and uji ≥ ci whenever qj > 0.

Corollary 1 shows that every Lindahl equilibrium allocation corresponds to the Nash

allocation for utility u(c) for some suitable c. We have interpreted the ci’s as a parameters

that modify consumers’ utilities. Alternatively, we can interpret the ci’s as modifying both

14

the utilities and the disagreement point in the bargaining game Bu. This will enable us to

relate Corollary 1 to Lemma 3 and Theorem 2.

Note that Lemma 3 and Theorem 2 together establish that if x is a Lindahl equilibrium

payoff for some utility v and x ∈ Bu for some u such that Bu ≤ Bv, then x is also a Lindahl

equilibrium payoff for u. A Lindahl allocation for u(c) yields payoffs η(Bu(c)). The scale

invariance property of the Nash Bargaining solution ensures that η(Bu(c))+c = η(Bu(c)+c).

Clearly, Bu ≤ Bu(c)+c and therefore η(Bu(c)+c) is a Lindahl equilibrium payoff of Bu

whenever it feasible for Bu, which is guaranteed if there is a Nash allocation q of Bu(c)+c

such that uji ≥ ci whenever qj > 0.

5. Applications

In this section, we analyze two applications of the Lindahl collective choice rule. In

the first, we consider one-to-one matching problems without transfers; in the second, we

consider general allocation problems with indivisible goods and no transfers.

5.1 Matching

A group of agents must decide who matches with whom as, for example, in the problem

of choosing roommates. A matching, is a bijection j from the set of all agents to itself such

that j(j(i)) = i for all i. If j(i) = i, then i is said to be unmatched. Some matchings may

be infeasible. For example, agents could be workers and firms and firms matching with

another firms may be disallowed. The set K = {1, . . . , k} represents the feasible matchings

and let Ni = {m| j(i) = m, for some j ∈ K} be the feasible matches of agent i and assume

that Ni 6= ∅ for all i.

Let wmi be the utility of agent i when she matches with agent m. We normalize

agents’ utilities so that being unmatched yields 0 utility for every agent and assume that

wmi > 0 for some m ∈ Ni. We eliminate all matchings that are not individually rational

and therefore, we assume wj(i)i ≥ 0 for all i and j. For notational convenience, we set

wmi = 0 if m 6∈ Ni.

We identify the following competitive market with this matching problem: each agent

has one unit of fiat money and agent i must pay price πmi for matching with agent m.

The price of remaining unmatched πii is zero for all agents. For notational convenience, we

15

also set πmi = 0 for m 6∈ Ni. Let πi = (π1i , . . . , π

ni ) be the corresponding price vector and

let π = (π1, . . . , πn). A commodity, in this setting, is an ordered pair (i,m) denoting i’s

match with m. Since we allow randomization, i’s consumption of (i,m) can vary between

0 and 1. Thus, i’s consumption set is:

Zi ={ξi ∈ IRn+

∣∣∣ ξi · e ≤ 1 and ξmi > 0 implies m ∈ Ni}

Then, given prices π, each agents solves the following maximization problem:

Wi(p) = maxξi∈Zi

wi · ξi subject to πi · ξi ≤ 1 (M1)

The consumption ξi is a minimum cost solution to the consumer’s problem if it solves the

above maximization problem and no other solution costs less than ξi at prices πi.

The firm’s revenue of matching j is the sum of the prices of the individual matches;

that is,∑ni=1 π

j(i)i . Therefore, the firm solves the maximization problem:

R(π) = maxq

∑j∈K

n∑i=1

πj(i)i · qj subject to q · e ≤ 1, qj ≥ 0 for all j (M2)

Let ξ = (ξ1, . . . , ξn) be a vector of consumer choices. The triple (π, ξ, q) is a (strong) Wal-

rasian equilibrium of the matching market v if ξi is a minimum cost solution to consumer

i’s problem (M1), q solves the firm’s maximization problem (M2) and for all i,m

ξmi =∑

{j| j(i)=m}

qj (M3)

Equation (M3) is the market clearing condition; that is, the requirement that consumer

i’s demand for matches with m are met by the firm’s supply of such matches.

The competitive economy differs from the collective choice market in the way con-

sumers express their demands. In the competitive economy, consumers specify demands

for private goods (partners m ∈ Ni) while in a collective choice market, consumers specify

their desired collective goods (matches j ∈ K). To map the matching market into a col-

lective choice market, define ui such that uji = wj(i)i . A Lindahl equilibrium (p, q) specifies

a distribution over matches q and a price pji that consumer i must pay for matching j.

16

Below, we relate Walrasian equilibria of the competitive economy to the Lindahl equilibria

of the collective choice market.

For any price p, let

πmi (p) =

{min{j:j(i)=m} p

ji if m ∈ Ni

0 otherwise

Let πi(p) be the corresponding price vector and note that it assigns each partner m ∈ Nithe price of the least cost match j that yields this partner. For any allocation q, define

ξi(q) such that ξmi (q) =∑{j:j(i)=m} q

j . Thus, ξi(q) is the partner assignment implied by

the allocation q.

Theorem 3: If (p, q) is a Lindahl equilibrium collective choice market, then (π(p), ξ(q), q)

is a Walrasian equilibrium of the competitive economy. If (π, ξ, q) is a Walrasian equilib-

rium of the competitive economy, then (p, q) such that pji = πj(i)i is a Lindahl equilibrium

of the collective choice market.

To see why the second part of Theorem 3 is true, note that every agent’s optimization

problem in the collective choice market is equivalent to their corresponding optimization

problem in the Walrasian setting. Condition (M3) implies that q achieves the desired

distribution of partners. Moreover, since ξi is a least cost solution for every consumer in

the competitive setting, so is q in the collective choice setting.

To see why the first part of Theorem 3 is true, note that in a Lindahl equilibrium

consumers choose minimum cost solutions to their optimization problems. Therefore,

j(i) = j′(i) and pji > pj′

i implies that qj = 0. It follows that ξmi (q) must be a least cost

solution to the i’s optimization problem at prices πi(p). For the same reason, the firm

revenue of any matching j such that qj > 0 is the same as in the collective choice market.

For any matching j such that qj = 0, the firm revenue is no greater than in the collective

choice market and, therefore, q must be optimal for the firm.

Since we have established the existence of a Lindahl equilibrium, Theorem 3 implies

that Walrasian equilibria of the matching market exist. Corollary 1, above, shows that the

Nash bargaining solution of the collective choice market can be implemented as a Lindahl

equilibrium and, therefore, Theorem 3 implies that the Nash bargaining solution can be

implemented as a Walrasian equilibrium of the competitive economy.

17

5.2 Allocating a Finite Set of Goods and Commodification

In the discrete good allocation problem, there is a finite set of goods H = {1, . . . , r}to be distributed to the n agents. There is no divisible good. Agent’s i utility for bundle

M is vi(M); we assume that vi(∅) = 0 and vi(L) ≤ vi(M) whenever L ⊂ M . We write

vi(h) instead of vi({h}). Let G(H) denote this general set of all such utility functions. A

utility function, vi, is additive if

vi(L ∪M) = vi(L) + vi(M)

whenever L∩M = ∅. The discrete allocation problem can be transformed into an exchange

economy by endowing each agent with one unit of fiat money and permitting random

consumption. Let Θi be the set of all possible random consumptions; that is, probability

distributions over 2H and let

P = {p : 2H → IR+ |p(∅) = 0}

be the set of all prices. Hence, we allow nonadditive (or combinatorial or package) prices.

The price p is additive if

p(L ∪M) = p(L) + p(M)

whenever L ∩M = ∅.Then, given any price p, a consumer’s budget is

B(p) :={θi ∈ Θi

∣∣∣ ∑M

p(M)θi(M) ≤ 1}

and her utility maximization problem is:

Vi = maxθi∈B(p)

∑M

vi(M)θi(M)

A random consumption θi is a minimal-cost solution to the utility maximization problem

if it is a solution to the maximization problem above and no other solution costs less.

Let A∗ be the set of all partitions of H. The firms maximization problem is:

R(p) = maxa∗∈A∗

∑M∈a∗

p(M)

18

A (feasible) allocation is a = (M1, . . . ,Mn) ⊂ Hr such that Mi ∩Ml = ∅ whenever i 6= l;

a random allocation is a probability distribution over allocations. For any allocation,

a = (M1, . . . ,Mn), ai denotes the i’th entry of a; that is, Mi. Let K be the set of all

allocations and Θ be the set of all random allocations. For any random allocation θ ∈ Θ,

let θi denote the i’th marginal of θ; that is,

θi(M) =∑

a:ai=M

θ(M)

The pair (p, θ) is a (strong) Walrasian equilibrium if for all i, θi is a minimal-cost

solution to consumer i’s utility maximization problem at prices p and∑i(p(ai)) = R(p)

for all a such that θ(a) > 0. It is easy to show that an arbitrary exchange economy,

v ∈ Gn(H) may have no Walrasian equilibrium in additive prices. Gul, Pesendorfer and

Zhang (2020) show that every gross substitutes economy v = (v1, . . . , vn) ∈ GSn(H) has a

Walrasian equilibrium with additive prices.5 Let W(v) be the set of Walrasian equilibrium

utility vectors for v.

The bargaining problem associated with the discrete allocation economy v is Bv :=

coco ({(v1(a), . . . , vn(a)) | a ∈ K} ∪ {o}). Clearly, each allocation problem yields a unique

bargaining problem; however, the converse is not true. Distinct allocation problems may

yield the same bargaining problem. Specifically, we may alter the property rights; that is,

the specification of the goods to be traded, without affecting the bargaining problem. In all

our formulations, the entire aggregate endowment initially belongs to the fictitious firm and

the n agents are all endowed with nothing other than a unit of fiat money. Nevertheless, the

set of Walrasian equilibrium payoffs depends on how the goods are defined. The example

below illustrates this point.

Office allocation example: Three agents must decide on an office allocation. One office

will be equipped with a premium desk, one with a good desk, and one with a standard desk.

We will offer two descriptions for this allocation problems that yield the same bargaining

problem but different Walrasian outcomes. The second description will yield all Lindahl

equilibria as a Walrasian equilibrium outcome.

5 Gross substitutes utilities are a class that includes additive utilities.

19

In the first version, the corresponding allocation problem has three goods and all

agents have unit-demands; that is, for M 6= ∅, we have vi(M) = maxh∈M vi(h) where

v1(1) = 10, v1(2) = 4, v1(3) = 2

v2(1) = 10, v2(2) = 7, v2(3) = 3

v3(1) = 10, v3(2) = 5, v3(3) = 1

Thus, good 1 is the office with the premium desk, good 2 is the office with the good desk

and good 3 is the office with the standard desk. This economy has a unique Walrasian

equilibrium in which good 2 is allocated to agent 2 and agents 1 and 3 each have an equal

chance at getting good 1 or 3. Hence, the equilibrium utilities are v1 = 6, v2 = 7, v3 = 5.5.

Consider an alternative description of the same situation. Each agent has a designated

office; that is, each agent can derive utility only if she gets her designated office; otherwise

her utility is 0. Each office is equipped with a standard desk. However, at most one desk

can be upgraded to a good desk and at most one desk can be updated to a premium desk.

Hence, this specification has 5 goods, the three offices (goods 1, 2 and 3) and the two

premium desks (good 4, the good desk and good 5, the premium desk). Each agent derives

utility both from her office and from any potential upgrade. The utility functions over the

relevant bundles are:

v1({1, 5}) = 10, v1({1, 4}) = 4, v1(1) = 2

v2({2, 5}) = 10, v2({2, 4}) = 7, v2(2) = 3

v3({3, 5}) = 10, v3({3, 4}) = 5, v3(3) = 1

This economy has many Walrasian equilibria and it can be shown, by appealing to Corollary

1, that the set of Walrasian equilibrium payoffs of this economy is the same its set of Lindahl

equilibrium payoffs. Notice that Bv = Bv and, thus, the two descriptions above represent

the same economy in terms of achievable utility profiles. This example show that it is

possible to commodify a bargaining problem B in two different ways (Bv = Bv = B)

leading to two distinct sets of Walrasian equilibria (W (v) 6= W (v)).

Theorem 4, below, generalizes the example. It shows that the set of Walrasian equi-

librium payoffs of any exchange economy are contained in the set of Lindahl equilibrium

20

payoffs of the corresponding collective choice market. Moreover, every bargaining prob-

lem can be commodified (with additive utilities if n = 2) so that the set of Walrasian

equilibrium payoffs with additive prices and the set of Lindahl equilibrium payoffs are the

same.

Theorem 4: (i) W(v) ⊂ L(v) for all v ∈ Gn(H). (ii) For all B ∈ Bo, there is v ∈ Gn(H)

such that Bv = B and W(v) = L(v); for n = 2, the preceding statement holds with additive

utilities and additive prices.

Theorem 4 and the example preceding it, reveal an advantage of Lindahl equilibria

over Walrasian equilibria. Lindahl equilibria depend only on the bargaining game while two

different exchange economies that yield the same bargaining game may have two different

sets of Walrasian equilibria. At the same time, Walrasian equilibria are simpler than

Lindahl equilibria because the former involve many fewer prices. This is so because the

number of allocations typically exceeds the number of goods and because Lindahl prices are

personal while Walrasian prices are not. However, if the commodity space is rich enough,

as is the commodity space we construct in the proof of Theorem 3, the distinction between

Lindahl equilibrium and Walrasian equilibrium disappears.

Note that commodification does not modify the firm’s technology. Though we al-

low non-linear pricing, we retain the assumption of an exchange economy. By contrast,

the standard mapping used in the proof of Lemma 1 introduces a new technology that

constrains the firm so that only feasible allocations can be “produced.”

6. Conclusion

To achieve equitable outcomes it is often necessary to randomize. In practice, this

randomization often occurs ex ante to give priority to agents while the mechanism remains

deterministic. For example, when allocating offices, an organization may first randomly

determine a priority order and then ask members to sequentially choose their preferred

office. As Hylland and Zeckhauser (1979) point out, such mechanisms lead to ex ante

inefficiency. As an alternative, they propose a market mechanism in which agents are

given a budget of fiat money and choose lotteries over the available offices. The Walrasian

mechanism proposed by Hylland and Zeckhauser is efficient and, if agents have identical

21

budget, equitable but limited in its applicability. Gul, Pesendorfer and Zhang (2019)

extend Hylland and Zeckhauser’s approach from unit demand to multi unit demand with

gross substitutes utilities. As we show in that paper, demand complementarities create

existence problems for the standard market mechanism. Moreover, externalities and public

goods render it inefficient. By contrast, the collective choice markets proposed in the

current paper are broadly applicable to all discrete allocation problems and always efficient.

In a collective choice market, each agent expresses her demand for social alternatives

rather than private outcomes. On the one hand, this allows us to deal with a much

broader range of applications but, on the other hand, the number of social alternatives can

be large and, therefore, the collective choice market may be too unwieldy to implement in

practice. For the case of matching markets, we show that Lindahl equilibria coincide with

standard competitive equilibria and, therefore, each agent need only consider the set of

possible partners and not the set of possible matchings (allocations) when formulating her

demand. An important direction for future research is to examine other circumstances in

which a smaller set of markets (and prices) suffices to implement Lindahl equilibria.

7. Appendix

7.1 Proofs of Lemmas 1-3 and Theorem 1

Proof of Lemma 1: Consider the following modified economy with private goods: let

X = {(x, β) = (x11, . . . , x

k1 , . . . , x

1n, . . . , x

kn, β) ∈ IRkn+1}

be the commodity space. There are n + 1 agents and every agent has endowment zero.

The consumption set of agent i, for i = 1, . . . , n, is

Di = {(x, β) ∈ X |xji ≥ 0 for all j, xjl = 0 for all l 6= i, β ≥ −1}

and her utility function is Ui(x, β) =∑kj=1 u

jixji . Agent n+ 1 has the consumption set

Dn+1 ={

(x, β) ∈ X∣∣∣ n∑j=1

minixji ≥ −1, β ≥ 0

}22

and her utility function is Un+1(x, β) = β.

An allocation (x(1), β1, . . . , x(n+ 1), βn+1) ∈ Xn+1 is feasible if (x(i), βi) ∈ Di for all

i,∑n+1i=1 x(i) ≤ 0 and

∑n+1i=1 βi ≤ 0. It is easy to verify that the set of feasible allocations

is compact, the consumption sets are convex and bounded below, and the utility functions

are quasi-concave, continuous and locally non-satiated. Let p be any price such that the

utility of every agent is bounded. Then, pβ > 0 and pji > 0 if uji > 0. It is straightforward

to verify that at any p that satisfies these conditions there exists, for every i = 1, . . . , n+1,

(x(i), βi), (x′(i), β′i) ∈ Di such that p · (x(i), βi) < p · (x′(i), β′i) ≤ 0. Thus, the conditions

of the equilibrium theorem (pg 12) in Gale and Mas-Colell (1975) are satisfied and the

economy has a Walrasian equilibrium. In that Walrasian equilibrium pβ > 0 and, therefore,

we can normalize prices so that pβ = 1. Local non-satiation of utilities implies that the

equilibrium allocation is Pareto efficient and that utility maximizing consumptions are

least-cost. Pareto efficiency implies that∑kj=1 mini x

ji (n+ 1) = −1.

If xjm(m) + mini xji (n + 1) < 0 for some m ∈ {1, . . . , n}, then utility maximiza-

tion of agent n + 1 implies pjm = 0 and, therefore, we can modify the allocation so that

xjm(m) = xji (i) for all j,m ∈ {1, . . . , n} and maintain all equilibrium conditions. Setting

q = (x11(1), . . . , xk1(1), pi) = (p1

i , . . . , pki ) then yields the desired Lindahl equilibrium.

Proof of Lemma 2: Let x ∈ ES(B). It follows that there is a simplex A with fair

solution x that B ≤ A and the fair solution of A is x. Since B ≤ A, x is Pareto efficient,

d(B) ≤ d(A) and b(B) ≤ b(A). Thus, d(B)+1/2(b(B)−d(B)) ≤ d(A)+1/2(b(A)−d(A)) = x.

For the converse, let x ∈ B be Pareto efficient and satisfy d(B) + 1/2(b(B) − d(B)) ≤ x.

Since E satisfies scale invariance, we may assume d(B) = o and b(B) = (1, 1). Since x

is Pareto efficient and B ∈ B, there is a = (a1, a2) such that a1, a2 > 0, a · x ≥ a · y for

all y ∈ B. Since x1, x2 > 0, we have ax > a1x1, ax > a2x2. Moreover, we must have

(2a2x2 ≥ ax ≥ 2a1x1) or (2a1x1 ≥ ax ≥ 2a2x2). Without loss of generality, assume the

latter holds. Let d1 = 2x1 − ax/a1 and note that d1 ∈ [0, x1). Then, let d = (d1, 0),

b = (a · (x−d)/a1, a · (x−d)/a2)+d and A = conv{d, (b1, 0), (d1, b2)}. It is straightforward

to verify that B ≤ A and F (A) = x and, therefore, x ∈ E(B) as desired.

23

Call A ∈ B a simplex if A = a×∆ + b for some a such that ai > 0 for all i and b such

that bi ≥ 0 for all i. We say that the simplex A supports B at x with (exterior normal) w

if x ∈ B ⊂ A, d(B) = d(A) and wy ≤ wx for all y ∈ A.

Lemma A1: (i) For all B there is a simplex A that supports B at η(B) with ∇fB(x).

(ii) For any simplex A, η(A) = 1nb(A) + n−1

n d(A).

Proof: The proof of (ii) is straightforward and omitted. Assume d(B) = o, let x = η(B)

and let w = ∇fB(η(B)) = (1/x1, . . . , 1/xn). Then, w · x = n and, since B is convex, it

follows that wy ≤ wx for all y ∈ B. Then, the simplex A = conv {o, nx1e1, . . . , nxne

n}

supports B at η(B) with ∇fB(η(B)).

For arbitrary B, let C = B − d(B). By the preceding argument, there is a simplex

A that supports C at η(C) with w = ∇fC(x). Then, note that η(B) = η(C) + d(B),

∇fB(η(B)) = ∇fC(η(C)) and therefore, A+ d(B) supports B at η(B) with ∇fB(η(B)).

Proof of Lemma 3: First, we show that N(B) ⊂ E(B): take x ∈ N(B). Hence, x ∈ B

and x = η(A) for some A such that B ≤ A. By Lemma A1, there is a simplex C that

supports A at x = η(A) with ∇Af(η(A)). Since C is a simplex, {x} = {η(A)} = F (C).

Hence, B ≤ A ≤ C and x ∈ F (C), therefore, x ∈ E(B).

Next, we show that E(B) ⊂ N(B). If x ∈ E(B), then there is A such that {x} = F (A).

Hence, x = η(A) and B ≤ A, therefore x ∈ N(B).

Proof of Theorem 1: That E satisfies scale invariance is obvious. If A and B are

simplices such that B ≤ A and F (A) ⊂ B, then A = B. Therefore E(A) = F (A) for every

simplex A. Since F (∆) = { 1ne}, it follows that E(∆) = { 1

ne}, as desired.

To prove consistency, let B ≤ A and x ∈ E(A) ∩ B. Then, there exists C such

that A ≤ C and F (C) = {x}. Hence, B ≤ C and therefore x ∈ E(B), as desired. To

prove justifiability, assume x ∈ E(B). Then, there is A such that B ≤ A and x ∈ F (A).

Hence, A is a simplex and therefore E(A) contains a single element x. That is, E satisfies

justifiability.

Next, assume that S satisfies the axioms. We will show that S = E. First, note

that scale invariance and symmetry imply that S(A) = F (A) for every simplex A. By

24

consistency, it then follows that x ∈ E(B) implies x ∈ S(B), and, therefore, E(B) ⊂ S(B)

for all B ∈ B. To prove that S(B) ⊂ E(B), let x ∈ S(B). Then, {x} = S(A) for some

A such that B ≤ A. Since E(A) ⊂ S(A), we must have E(A) = {x} and since E satisfies

consistency, x ∈ E(B) as desired.

7.2 Proofs of Theorem 2 and Corollary 1

Let p be a price. In a collective choice market the consumer solves

Ui(p) = maxqui · q subject to pi · q ≤ 1, e · q ≤ 1 (P)

The dual of problem (P) is

minµ0,µ1≥0

µ0 + µ1 subject to µ0e+ µ1pi ≥ ui (D)

The vector (q, µ0, µ1) is feasible if q satisfies the constraints of (P) and (µ1, µ2) satisfies the

constraints of (D). A feasible vector (q, µ0, µ1) is optimal (that is, q solves (P) and µ0, µ1

solves (D)) if and only if

µ0(q · e− 1) = 0

µ1(q · pi − 1) = 0

for all j, qj(µ0 + µ1pji − uji ) = 0

(CS)

Let J(q) = {j | qj > 0}. For any utility ui and ci < maxj uji , let uji (ci) = max{0, uji − ci}.

Note that ui(ci) is a utility.

Lemma A2: Let q ∈ Q be a minimal cost solution to consumer i’s maximization problem

for utility ui and prices p such that q · pi = 1. There are c ≥ 0 and α > 0 such that

αpji ≥ uji − c for all j and αpji = uji − c for j ∈ J(q). Moreover, q is a minimal cost solution

to consumer i’s maximization problem for utility ui(c) and prices p.

Proof: Let µ0, µ1 be the associated solution of the dual (D). First, consider the case in

which uji 6= umi for some j,m ∈ J(q). Then, (CS) implies µ1 > 0. Set c = µ0, α = µ1.

Feasibility and (CS) imply that αpji ≥ uji − c with equality if j ∈ J(q), as desired.

25

It remains to show that q is a minimal cost solution to the consumer’s maximization

problem given utility ui(c). Since (q, µ0, µ1) is feasible (for utility ui), we have αpji ≥ uji−c

for all j. We also have αpji ≥ ui(c) for all j, since the left-hand side is non-negative.

The first part of the Lemma implies uji − c = uji (c) for all j ∈ J(q) and, therefore,

qj(αpji − ui(c)) = 0 for all j. Hence, (q, µ0, µ1) such that µ0 = 0, µ1 = α is a feasible

solution for (P) and (D) that satisfies (CS) for utility u(c); that is, q solves consumer i’s

maximization problem given utility uji (c). Since uji 6= umi for some j,m ∈ J(q), we have

uji (c) = uji − c 6= umi − c = umi (c) and therefore, this solution must be minimal cost.

Second, consider the case where β = uji = umi for all j,m ∈ J(q). Since q is a minimal

cost solution and q ·pi = 1, we must have pji = pmi = 1 for all j,m ∈ J(q). Let µ0, µ1 be the

associated solution of the dual (D). If µ1 > 0, set α = µ1 and repeat the argument above

to conclude that q is a solution to (P) given utility ui(c) and uji (c) = uji − c = αpji = α > 0

for all j ∈ J(q). Then, since q is a minimal cost solution for ui implies q is a minimal cost

solution for ui(c).

If µ1 = 0, set c = max{j|pji<1} u

ji if {j | pj < 1} 6= ∅ and c = 0 otherwise and set

α = β − c. Note that β − c > 0 since q is a minimal cost solution. Then, c+ αpji ≥ c ≥ uji

if pji < 1 and c + αpji ≥ µ0 ≥ uji if pji ≥ 1. Therefore, αpji ≥ uji − c for all j and with

equality if j ∈ J(q). Then, note that (q, c, α) is a feasible vector that satisfies (CS) and

hence q is a solution to the consumer’s optimization problem. Since β − c > 0 and q is a

minimal cost solution to the optimization problem for ui, it must also be a minimal cost

solution for ui(c).

Lemma A3: If there is λ = (λ1, . . . , λn) ≥ o such that vi = ui + (λi, . . . , λi) for all i,

then L(u) + λ ⊂ L(v).

Proof: Suppose such a λ exists and choose x ∈ L(u) and let (p, q) be a Lindahl equilibrium

of u that yields the payoff vector x. Then, for all i and q such that e · q ≤ 1, ui · q ≤ ui · q =

vi · q−λi and vi · q−λi ≤ ui · q. It follows that q is a minimal cost solution to consumer i’s

maximization problem in v. Clearly, q is a solution to the firm’s maximization problem in

the collective choice market v and hence (p, q) is a Lindahl equilibrium of v and therefore,

x+ λ ∈ L(v).

26

Proof of Theorem 2: First, we show that x ∈ E(Bu) implies x ∈ L(u). Let x ∈ E(Bu).

Then, there is a simplex A = na⊗∆ + d such that Bu ≤ A and x = a+ d ∈ Bu. Assume

for now, that d = o so that d(A) = d(Bu) = o. Then, Bu ⊂ A. For any j ∈ K, let

yj = (uj1, . . . , ujn). Since yj ∈ Bu ⊂ A, it is a unique convex combination of the extreme

points of A. Let zji be the weight of naiei in that convex combination and set pji = nzji .

Note that uji > 0 if and only if pji > 0 and that uji/pji = ai if pji > 0. Since x ∈ Bu, it is a

convex combination of the extreme points of Bu. Let qj be the weight of yj in that convex

combination. By construction, ui · q = ai. Note that for any q such that q · pi ≤ 1 we have

ui · q =

K∑j=1

aipji · q

j ≤ ai

with equality only if q · pi = 1. Therefore, q is a minimal cost solution to the consumer’s

problem and (p, q) is a Lindahl equilibrium.

If o = d(Bu) 6= d; that is, d(Bu) ≤ d(A), then repeat the above construction for

A′ = A − d to show that x − d is a Lindahl equilibrium payoff vector of the economy v

such that vji = uji − di. Then, Lemma A3 ensures that x is a Lindahl equilibrium payoff

vector for the economy u.

Next, we show that x ∈ L(u) implies x ∈ E(Bu). Let x ∈ L(u) and let (p, q) be the

corresponding Lindahl equilibrium outcome. First, consider the case in which pi · q = 1 for

all i. By Lemma A2, there is, for each i, some ci ≥ 0 and αi > 0 such that uji − ci ≥ αipji

for all j with equality for j ∈ J(q). Furthermore, Lemma A2 implies that (p, q) is also a

Lindahl equilibrium for the collective choice market (ui(ci))ni=1. Since uji (ci) = αip

ji for all

j ∈ J(q) and pi · q = 1, we have ui(ci) · q = αipi · q = αi and xi = ci + αi. Let z ∈ Bu and

let q be such that zi = ui · q.

Since αipji ≥ u

ji − ci for all i, j, we have αipi · q ≥ zi− ci for all i or pi · q ≥ (zi− c)/αi.

Note that the firm’s profit cannot be greater than n, the aggregate endowment of money.

Therefore, firm optimality implies n ≥∑ni=1 pi · q ≥

∑ni=1 pi · q. Thus, we obtain

n ≥n∑i=1

zi − cαi

(1)

27

for all zi ∈ Bu. Let d = (c1, . . . , cn), a = (α1, . . . , αn) and A = na⊗∆ + d and note that

(1) implies A ≥ Bu. Clearly, F (A) = a+ d = x and, therefore, x ∈ E(Bu).

Finally, consider the case in which pi · q < 1 for some i. Let I be the set of all such

agents. Let J∗i = {j|uji ≥ umi ∀m} be the bliss outcomes for i. If pi · q < 1 for i ∈ I, then

J(q) ⊂ J∗i since otherwise i is not choosing a utility maximizing plan. Furthermore, since

q is a minimal cost solution for consumer i ∈ I, pji = pmi for all i ∈ I and j,m ∈ Ji(q).

Define p = (p1, . . . , pn) as follows: pji = 1 if i ∈ I and j ∈ Ji and pji = pji otherwise.

It is easy to see that (p, q) is a Lindahl equilibrium. Moreover, every consumer satisfies

pi · q = 1 for all i. Thus, we can apply the argument above to show that x ∈ E(Bu).

Proof of Corollary 1: For c ∈ C(u), define Hi(ci) = conv{ej |uji ≥ ci} and note

that uji (c) · q + c = uji · q if and only if q ∈ Hi(ci). Let H(c) =⋂ni=1Hi(ci). Let q be the

Nash allocation for u(c).

To prove the first statement of the corollary, q is a Nash allocation for u(c), c ∈ C(u)

and uji ≥ ci whenever qj > 0. Hence, q ∈ H(c) and for all i, there is j such that uji (ci) > 0.

Therefore, the constraint q · e ≤ 1 is binding. Then, the first order condition of (N) yields

λ > 0 such thatn∑i=1

uji (ci)

ui(ci) · q≤ λ (F )

andn∑i=1

uji (ci)qj

ui(ci) · q= λqj

Then, q · e = 1 implies λ = n. Set µ0i = ci, µ

1i = ui(ci) · q and pi = ui(ci)/(ui(ci) · q).

Hence, pi · q = 1 and q ∈ Hi(ci) and therefore, µ0i +µ1

i pji ≥ u

ji for all j. Moreover, equality

holds for all j such that qj > 0. That is, (µi0, µi1, q) is a feasible solution that satisfies

(CS) and, therefore, is optimal. It remains to show that q solves the firm’s problem at

prices (p1, . . . , pn). Since λ = n, inequality (F) implies that∑ni=1 p

jieji ≤ n for all j and,

therefore, firm profit cannot exceed n. Since q yields profit n, it maximizes profit.

To prove the second statement of the corollary, assume that (p, q) is a Lindahl equi-

librium. Then, consumer optimality implies that there are µi0, µi1 such that µ0

i +µ1i pji ≥ u

ji

28

for all j where equality holds for all j such that qj > 0. Therefore, µ0i + µ1

i = ui · q and

µ1i ≥ u

ji (µ

i0) for all j with equality if qj > 0. Summing over all i we obtain

n∑i=1

uji (ci)

µi=

n∑i=1

uji (ci)

ui(µi0) · q≤ n (F )

with equality if qj > 0. Thus, q is the Nash allocation for u(c). Moreover, if ej 6∈ Hi(µ0i )

then µi0 + µi1pji ≥ µi0 > uji and, therefore, qj = 0. Hence q ∈ Hi(µ

0i ) = Hi(ci) for all i.

7.3 Proofs of Theorems 3 and 4

Proof of Theorem 3: Lemma A1(i) yields a simplex A = conv {z1e1, . . . , zne

n} (zi > 0

for all i) that supports Bw at η(Bw) with ∇fBw(η(Bw)). Hence, η(A) = η(Bw) and

d(B) = d(A) = o and therefore, by Lemma A1(ii), η(Bw) = 1ne ⊗ z. Let q be any

probability distribution over K such that∑j q

j(wj1, . . . , wjn) = η(Bw).

Each x ∈ Bw and is a unique convex combination of the extreme point of A. That

is, xi = λi · zi for λ1, . . . , λn ≥ 0 such that∑i λi ≤ 1. Let λi(x) be the weight of zi

in that convex combination. For x = (wj1, . . . , wjn), let pji = λi(w

ji ) and let πmi = pji for

some j such that j(i) = m. Clearly, λ(wji ) = λ(wj′

i ) whenever j(i) = j′(i). Hence, πmi is

well-defined. Verifying that (π, q) is a WE and (p, q) is a LE is straightforward.

To conclude the proof, we will show that any WE is an LE. Let (π, q) be an WE and

let pji = πj(m)i . Then, clearly, (p, q) is a WE.

Proof of Theorem 4: Let (p, θ) be a WE and let K = {a(1), . . . , a(k)} be the set of all

feasible allocations. Then, for every feasible allocation a(j), let

pji =∑h∈ai

ph

Then, it is easy to verify that (p, θ) is an LE.

To prove the second assertion, let n = 2 and y1 = (y11 , y

12), . . . , yk = (yk1 , y

k2 ) be all of

the vertices on the Pareto frontier of B ∈ Bo ordered so that y11 < y2

1 < . . . < yk1 . First

consider the case in which y11 = 0 and yk2 = 0. Then, let H = {1, . . . , k − 1}, v1(h) =

yh+11 − yh1 and v2(h) = yh2 − yh+1

2 for h ∈ H. For ∅ 6= M ⊂ H, let vi(M) =∑h∈M ui(h)

for i = 1, 2. Hence, vi is additive.

29

Since f(h) =yh2−y

h+12

yh+11 −yh1

is an increasing function, in any efficient allocation of the

economy v, there is l such that agent 2 consumes M1(l) = {1, . . . , l} and agent 1 consumes

M2(l) = {l + 1, . . . k − 1}. Hence, Bv = B. A random allocation is Pareto efficient if

and only if it has at most two elements in its support, and hence is of the form a(l) =

(a1(l), a2(l)) = (M1(l),M2(l)) with probability γ ∈ (0, 1] and a(l + 1) = (a1(l + 1), a2(l +

1)) = (M1(l + 1),M2(l + 1)) with probability 1− γ.

Let x ∈ Bv be a Lindahl equilibrium payoff vector. Hence, ql the probability of

the allocation a(l), is γ and ql+1 = 1 − γ for a(l) and a(l + 1) as above. By Lemma 2,

x ≥ b(Bv)/2; in particular, x1 ≥ yk1/2. Since x is Pareto efficient, there is a ∈ IR2++ such

that ax ≥ az for all z ∈ Bv. We must have a2x2 ≥ a1x1 or a1x1 ≥ a2x2. Without loss of

generality, assume the latter holds. Then, let ph = v1(h)/x1 for all h ≤ l if γ = 1 and for

all h ≤ l+ 1 (if γ < 1). Hence, the random consumption that yields a1(l) with probability

γ and a1(l + 1) with probability 1 − γ is just affordable for agent 1. Moreover, agent 1’s

util per price of the goods she is to consume is constant at 1/x1.

Note that since x1 ≥ yk1/2, if we were to set the price, ph, of each of the remaining

goods to v1(h)/x1, agent 2 could purchase the random consumption that yields a2(l) with

probability γ and a2(l + 1) for less than 1, her endowment of fiat money.

Conversely, if we were to set ph for each of the remaining goods to a2v2(h)/a1x1, agent

2 would not be able to afford the random consumption that yields a2(l) with probability

γ and a2(l + 1) if a2x2 > a1x1. Let Ih =[v1(h)x1

, a2v2(h)a1x1

]. Since a2/a1 is the slope of the

tangent line through the point x, Ih 6= ∅ for all h > l+ 1 if γ < 0 and for all h > l if γ = 1.

Note also that if γ < 1, the interval I l+1 is a single point. Hence, we can choose ph ∈ Ih

for all such h ≥ l of γ < 1 and for all h ≥ l+ 1 if γ = 1 so that agent 2 can just afford the

random consumption (γ, a2(l); 1 − γ, a2(l + 1)). Moreover, agent 2’s util per price of the

goods he consumes in no greater than a2/a1x1.

Since the slope of f is increasing, agent 1’s util per price for goods h ≥ l+ 1 is greater

than 1/x1 while agent 2’s util per price of goods h ≤ l is greater than a2v2/a1x1. It follows

that (γ, ai(l); 1 − γ, ai(l + 1)) is a minimal cost solution to the maximization problem of

agent i. Hence, x ∈W (v).

If yk−12 > 0, then let H = {1, . . . , k} and define the utility of the new good k as follows:

v1(k) = 0 and v2(k) = yk−12 . If y1

1 > 0, then add a new good 0, (or another new good 0,

30

if k has already been added) to H such that v1(0) = y11 and v2(0) = 0. Then, repeat the

construction above, after extending each vi to all subsets of H additively to derive, once

again, a WE that yields x as its payoffs.

For the general case; that is, for n ≥ 2 and any polytope B ∈ Bo, we define the

commodity space as follows: let X be the set of Pareto efficient extreme points of B. Then,

let Xi be the projection of X to the i’the coordinate; that is, Xi = {xi |x ∈ X} ∪ {0} and

define the set of possible partial payoff vectors as

X0 = (X1 ∪ {−1})× (X2 ∪ {−1})× · · · × (Xn ∪ {−1})

For y ∈ X0, xi ∈ Xi let (y−i, xi) ∈ X0 be the vector in which the i−coordinate of y is

replaced by xi. Define the set of consistent partial payoff vectors and minimally inconsistent

partial payoff (MIPP) vectors as follows:

X+ = {x | there is y ∈ X such that xi ∈ {−1, yi} for all i }

X− = {x ∈ X0\X+ | there is i such that (x−i,−1) ∈ X+}

For any x ∈ X−, let δ(x) be the cardinality of the set {i |xi 6= −1} minus 1 and let

Y (x) = {(x, 1), . . . , (x, δ(x))} be a set that contains exactly δ(x) copies of x. (If δ(x) = 0

then Y (x) = ∅.) Let Yi = {(x, i)|x ∈ Xi} and define the set of goods as follows:

H =⋃i

Yi ∪⋃

x∈X−Y (x)

Hence, H contains one copy of every possible payoff for each agent i and δ(x) copies of

every MIPP x ∈ X−; that is, one fewer copy than the number of possible payoffs in the

MIPP vector x (i.e., entries that are not −1). Then, for any M ⊂ H, define the utility of

agent i for the bundle M as follows:

vi(M) = max{y | gi(M,y) = 1}

where gi(M,y) = 1 if the following two conditions hold:

(i) (y, i) ∈M and there is no y ≤ y such that (y, i) ∈ Yi\M ;

(ii) Y (x) ∩M 6= ∅ for all x ∈ X− such that xi = y.

31

Next, we will show that Bv = B. Take y ∈ X and for every i and x ∈ X− such that

xi = yi, choose a distinct (x, ki) ∈ Y (x). Then, let

Mi = {(y, i) ∈ Yi|y ≤ y} ∪ {(x, ki)|x ∈ X−1, xi = yi}

By construction, v(Mi) = yi. To see that this collection of Mi’s, for i = 1, . . . , n, is feasible,

note that if such a collection were not feasible, there would be some x ∈ X− such that the

cardinality of the set {i |xi = yi} is greater than the cardinality of the set Y (x). But this

contradicts the definition of δ(x).

Thus we have shown X ⊂ Bv and therefore coco (X ∪ {o}) = B ⊂ Bv. To see that

Bv ⊂ B, take any collection of disjoint subsets M1, . . .Mn,⊂ H. Let yi = vi(Mi) for all i

and assume that y = (y1, . . . , yn) /∈ B. It follows that there is x ∈ X− such that xi = yi

for all i such that xi 6= −1. Then, each Mi must contain an element of Y (x), contradicting

the definition of δ(x).

Finally, we will show that L(v) ⊂ W (v). Let K be the set of all feasible allocations

of goods for the competitive economy v. For any allocation a ∈ K, define uai = vi(ai). Let

q be a Lindahl equilibrium outcome let p be the Lindahl equilibrium price. By Corollary

1,we may choose the Lindahl equilibrium price such that pji = pli whenever consumer i

receives the same bundle in allocation j and l. Next, we define the price p: we say that

M ⊂ H is minimal if (for some ai) it is a minimal subset of ai such that vi(M) = vi(ai).

Let p(M) = pai if M is minimal. For all other M ⊂ H, define p(M) = max{p(L)|L ⊂

M,L minimal}.

Let θi(q) be the random consumption of consumer i implied by the Lindahl equilibrium

allocation q in the collective choice market and let θ(q) = (θ1(q), · · · θn(q)). Note that

consumer i’s utility from q is the same as the utility from θi(q) and, moreover, the two

cost the same. Moreover, for every random consumption θi in the competitive economy,

there is q that yields the same utility and costs the same. Therefore, the minimal-cost

optimality of θi(q) follows from the minimal cost optimality of q.

It remains to show optimality for the firm. Let {M1, . . . ,M l} ∈ A∗ be any partition

of H. From the definition of vi it follows that k 6= m implies vi(Mk) · vi(Mm) = 0 for all i.

Hence, for every i there is at most one ki ∈ {1, . . . , l} such that vi(Mki) > 0. For Mk such

32

that vi(Mk) = 0 for all i we must have p(Mk) = 0. Thus, if

∑lk=1 p(M l) > R(p) it follows

that there is an allocation a such that∑lk=1 p(M l) =

∑ni=1 p

ai . Since this contradicts

the firm-optimality of (p, q) in the Lindahl equilibrium it follows that θ(q) maximizes the

firm’s profit.

33

References

Brandl, F., F. Brandt, D. Peters, C. Stricker, W. Suksompong (2020), “Funding PublicProjects: A Case for the Nash Product Rule,” working paper.

E. Eisenberg and D. Gale (1959), “Consensus of subjective probabilities: The pari-mutuelmethod.” Annals of Mathematical Statistics, 30(1):165168.

B. Fain, A. Goel, and K. Munagala (2016), “The core of the participatory budgeting prob-lem,” In: Proceedings of the 12th International Conference on Web and Internet Economics(WINE), Lecture Notes in Computer Science (LNCS), pages 384399, 2016.

Fujishige, S. and Z. Yang (2003) “A Note on Kelso and Crawford’s Gross SubstitutesCondition,” Mathematics of Operations Research, 28, 463–469.

Foley, D. (1967) Resource Allocation and the Public Sector, Yale Economic Essays, 3(Spring, 1967)

Hyland, A. and R. J. Zeckhauser (1979) “The Efficient Allocation of Individuals to Posi-tions,” Journal of Political Economy, 87, 293–314.

Kalai, E. and M. Smorodinsky (1975) “Other Solutions to Nash’s Bargaining Problem,”Econometrica, 43, 513–518.

Kelso, A. S. and V. P. Crawford (1982) “Job Matching, Coalition Formation, and GrossSubstitutes,” Econometrica, 50, 1483–504.

Lindahl, E. (1919) Just Taxation - A Positive Solution. Reprinted in part in: R.A. Musgraveand A. Peacock, eds., Classics in the theory of public finance (Macmillan, London, 1958).

Mas-Colell, A., (1992) “Equilibrium Theory with Possibly Satiated Preferences,” Equilib-rium and Dynamics: Essays in Honour of David Gale, edited by M. Majumdar, London,Macmillan, 201–213.

Thomson, W, (1994) “Cooperative Models of Bargaining,” in: Handbook of Game Theory,Volume 2, Edited by R.J. Aumann and S. Hart, Elsevier Science B.V., 1238–1284.

Nash, J.F. (1950) “The Bargaining Problem,” Econometrica, 28: 155–162.

Perles, M.A. and M. Maschler (1981) “A Super-Additive Solution for the Nash BargainingGame,” International Journal of Game Theory, 10: 163-193.

D. Schmeidler and K. Vind, (1972), “Fair Net Trades,” Econometrica 40, 637–642.

Hal R. Varian (1973) “Equity, Envy, and Efficiency” Journal of Economic Theory, 9, 63–91

34